Matches in SemOpenAlex for { <https://semopenalex.org/work/W4319996489> ?p ?o ?g. }
- W4319996489 endingPage "4026" @default.
- W4319996489 startingPage "4011" @default.
- W4319996489 abstract "The one-stage object detector has recently attracted extensive interest due to its high detection efficiency and simple framework. However, one-stage detectors suffer much from the extreme positive-negative imbalance since the majority of samples are labeled as the negative. To address this problem, a novel activation function, the Gradient Enhanced Function (GEF), is devised for the last output layer of the one-stage detector to strengthen the role of positive samples during the optimization process. The Quality-Guided Loss (QGL) is proposed for the classification task in object detection, which further makes the training focus on the high-quality positive samples. In addition, QGL is devised to handle different classification labels (i.e. the one-hot label and the soft label). Owing to its simplicity and effectiveness, the QGL together with the GEF might be applicable to various one-stage methods, including anchor-based and anchor-free detectors. Comprehensive experiments are conducted on multiple public benchmarks, MS COCO, PASCAL VOC and Berkeley DeepDrive. The results demonstrate that the proposed approach achieves notable improvements in both anchor-based and anchor-free detectors with various classification labels. The effectiveness of the GEF and the QGL is further verified in the stronger one-stage detectors." @default.
- W4319996489 created "2023-02-11" @default.
- W4319996489 creator A5002536027 @default.
- W4319996489 creator A5020535366 @default.
- W4319996489 creator A5040153479 @default.
- W4319996489 creator A5051083091 @default.
- W4319996489 date "2023-08-01" @default.
- W4319996489 modified "2023-09-29" @default.
- W4319996489 title "Balanced One-Stage Object Detection by Enhancing the Effect of Positive Samples" @default.
- W4319996489 cites W1536680647 @default.
- W4319996489 cites W2031489346 @default.
- W4319996489 cites W2088049833 @default.
- W4319996489 cites W2108598243 @default.
- W4319996489 cites W2183182206 @default.
- W4319996489 cites W2194775991 @default.
- W4319996489 cites W2344484090 @default.
- W4319996489 cites W2549139847 @default.
- W4319996489 cites W2565639579 @default.
- W4319996489 cites W2570343428 @default.
- W4319996489 cites W2607037079 @default.
- W4319996489 cites W2799058067 @default.
- W4319996489 cites W2808989867 @default.
- W4319996489 cites W2902609089 @default.
- W4319996489 cites W2906580494 @default.
- W4319996489 cites W2907183510 @default.
- W4319996489 cites W2917092108 @default.
- W4319996489 cites W2928165649 @default.
- W4319996489 cites W2962721361 @default.
- W4319996489 cites W2963037989 @default.
- W4319996489 cites W2963113370 @default.
- W4319996489 cites W2963299996 @default.
- W4319996489 cites W2963351448 @default.
- W4319996489 cites W2963516811 @default.
- W4319996489 cites W2963604034 @default.
- W4319996489 cites W2963786238 @default.
- W4319996489 cites W2963813458 @default.
- W4319996489 cites W2964093967 @default.
- W4319996489 cites W2964444661 @default.
- W4319996489 cites W2982770724 @default.
- W4319996489 cites W2984009799 @default.
- W4319996489 cites W2988452521 @default.
- W4319996489 cites W2989676862 @default.
- W4319996489 cites W3010134323 @default.
- W4319996489 cites W3015352266 @default.
- W4319996489 cites W3016916515 @default.
- W4319996489 cites W3020833432 @default.
- W4319996489 cites W3035134467 @default.
- W4319996489 cites W3035396860 @default.
- W4319996489 cites W3035473155 @default.
- W4319996489 cites W3035524459 @default.
- W4319996489 cites W3035564946 @default.
- W4319996489 cites W3109381875 @default.
- W4319996489 cites W3110525903 @default.
- W4319996489 cites W3117854388 @default.
- W4319996489 cites W3152635971 @default.
- W4319996489 cites W3163511872 @default.
- W4319996489 cites W3172087149 @default.
- W4319996489 cites W3172971995 @default.
- W4319996489 cites W3176187859 @default.
- W4319996489 cites W3203170887 @default.
- W4319996489 cites W4206487945 @default.
- W4319996489 cites W4214494681 @default.
- W4319996489 doi "https://doi.org/10.1109/tcsvt.2023.3237826" @default.
- W4319996489 hasPublicationYear "2023" @default.
- W4319996489 type Work @default.
- W4319996489 citedByCount "0" @default.
- W4319996489 crossrefType "journal-article" @default.
- W4319996489 hasAuthorship W4319996489A5002536027 @default.
- W4319996489 hasAuthorship W4319996489A5020535366 @default.
- W4319996489 hasAuthorship W4319996489A5040153479 @default.
- W4319996489 hasAuthorship W4319996489A5051083091 @default.
- W4319996489 hasConcept C111919701 @default.
- W4319996489 hasConcept C127413603 @default.
- W4319996489 hasConcept C146357865 @default.
- W4319996489 hasConcept C151730666 @default.
- W4319996489 hasConcept C153180895 @default.
- W4319996489 hasConcept C154945302 @default.
- W4319996489 hasConcept C199360897 @default.
- W4319996489 hasConcept C201995342 @default.
- W4319996489 hasConcept C2776151529 @default.
- W4319996489 hasConcept C2780451532 @default.
- W4319996489 hasConcept C2781238097 @default.
- W4319996489 hasConcept C41008148 @default.
- W4319996489 hasConcept C75608658 @default.
- W4319996489 hasConcept C76155785 @default.
- W4319996489 hasConcept C86803240 @default.
- W4319996489 hasConcept C94915269 @default.
- W4319996489 hasConcept C98045186 @default.
- W4319996489 hasConceptScore W4319996489C111919701 @default.
- W4319996489 hasConceptScore W4319996489C127413603 @default.
- W4319996489 hasConceptScore W4319996489C146357865 @default.
- W4319996489 hasConceptScore W4319996489C151730666 @default.
- W4319996489 hasConceptScore W4319996489C153180895 @default.
- W4319996489 hasConceptScore W4319996489C154945302 @default.
- W4319996489 hasConceptScore W4319996489C199360897 @default.
- W4319996489 hasConceptScore W4319996489C201995342 @default.
- W4319996489 hasConceptScore W4319996489C2776151529 @default.
- W4319996489 hasConceptScore W4319996489C2780451532 @default.